The objective of this research is to identify how visual impairment impacts the safety and performance of older drivers. Driving cessation in older adults is tied to many negative health indicators including increased risk for depression, reduced access to healthcare services, and increased likelihood of placement into long-term-care. The societal significance of research on older drivers is dramatically apparent when we recognize that vision impairment is common in the older adult population, and that older drivers have among the highest rates of motor vehicle collisions compared to other age groups. A critical scientific challenge is to identify what aspects of vision are required for safe driving by older adults. Previous research on older driver safety and performance has focused on the study of police accident reports, brief driving evaluations, or driving simulator studies. These techniques have serious scientific limitations; crash studies do not provide information about what lead up to the collision; in on-road studies research personnel are in the vehicle potentially biasing how the driver behaves; simulator studies present artificial, cartoon-like scenarios that over-simplify the roadway environment. All these issues undermine the ability to understand the actual visual factors underlying driver performance and safety. Naturalistic driving methods overcome these limitations by unobtrusively placing multiple sensors and video cameras in the vehicle that record vehicle kinematics, driver behaviors (e.g., gaze behavior), and roadway environment. They allow driving to be studied in an objective way over lengthy time periods (months) in the driver's own vehicle where the driver chooses the route during the course of everyday life. In this research we will use naturalistic driving techniques to examine the relationships between vision and older driver safety and performance. We focus on older drivers with normal vision and those with contrast sensitivity deficits and slowed visual processing speed, since these problems are the strongest and most consistently identified visual risk factors for driving problems in the older adult population. Analyses will focus on the relationship between vision and critical safety events (crashes, near-crashes), lane-keeping, turning at intersections, driving under secondary tasks demands, and when a co-pilot (passenger) is present. These relationships will be examined in light of other driver characteristics, environmental factors, and vehicle factors. Driving performance as measured by naturalistic driving methods will also be compared to driving ratings provided by a driving rehabilitation specialist on a standardized driving route, which is currently the clinical gold standard. This research will stimulate the development of interventions to enhance driving mobility in visually impaired older adults and thus improve quality of life.